MAINT: Refactor _page.py

[MartinThoma]

This is an experiment to see how well Github Copilot works

The goal of this PR is to make the PageObject easier to understand and maintain.

For the Reviewer

• I'm uncertain if the license header is necessary / correct. I think most (all?) of the text extraction part is written by @pubpub-zz
• See Refactor regular text extraction into dedicated module #3010

[codecov]

Codecov Report

Attention: Patch coverage is 95.72650% with 10 lines in your changes missing coverage. Please review.

Project coverage is 96.71%. Comparing base (af645a4) to head (347bdf4).

Files with missing lines	Patch %	Lines
pypdf/_text_extraction/_text_extractor.py	95.65%	7 Missing and 3 partials ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3339      +/-   ##
==========================================
- Coverage   96.73%   96.71%   -0.03%     
==========================================
  Files          54       54              
  Lines        9073     9096      +23     
  Branches     1676     1661      -15     
==========================================
+ Hits         8777     8797      +20     
- Misses        177      180       +3     
  Partials      119      119              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[stefan6419846]

Please re-use the existing text extraction submodule instead of a new one. Additionally, we should review the coverage.

Just for interest: Did you do a complete review/diff of the changes to see whether anything was changed during refactoring without it being expected?

[MartinThoma]

Just for interest: Did you do a complete review/diff of the changes to see whether anything was changed during refactoring without it being expected?

I did some smoke-tests and all parts I checked were identical.
(Except for the parts that I wanted to be changed, of course 😄 )

However, I did not check everything.

[MartinThoma]

I was pleasently surprised how well agent mode of Github copilot worked for this. I essentially gave it #3010 (comment) (rephrased a little bit) and it did almost do all the work. After that I used ruff check / ruff format + fix some mypy issues and that was it :-)

[MartinThoma]

Additionally, we should review the coverage.

The missing lines are not new. So I would expect that they are currently missing as well.

[MartinThoma]

Please re-use the existing text extraction submodule instead of a new one.

I'll look into it 👍

[MartinThoma]

@stefan6419846 I feel like this change is too hard to review. What do you think about me doing this in 2 (or more) PRs?

In a first PR, I would create pypdf/_text_extraction/_text_extractor.py: I would create the TextExtraction class with a basic initializer that gets called from PageObject._extract_text, but not refactor how _extract_text works internally. So most of the code there should be character-for-character identical.

[stefan6419846]

@MartinThoma The easier to review the code, the better - there surely are similar reasons why some of the larger recent PRs are still in my review queue.

[stefan6419846]

@MartinThoma What is the current state of this PR?